The proposed research will follow-up an established cohort of about 1,000 HIV seronegative intravenous drug users (IVDUs) in Baltimore at six month intervals in order to identify determinants and temporal trends in HIV incidence. At each 6 month visit, we will obtain blood specimens and detailed behavioral interviews. Medical and death records will be accessed for morbidity and mortality data. The interview data will allow the identification of factors associated with behavioral change in HIV- seronegative IVDUs over time, including reduction in unsafe behaviors, adoption of safe behaviors and relapse. The data on morbidity and mortality in a cohort of HIV seronegative IV drug users, will permit comparisons of the experience of seronegatives with a cohort of HIV seropositive IV drug users already being followed in the ALIVE study (DA04334) using the same laboratory, clinical setting, and interview schedule. By providing comparative, cause-specific morbidity and mortality among HIV seronegatives, this study will provide essential data to determine which non-AIDS illnesses are HIV-related among IV drug users. Our longitudinal study of the natural history of drug abuse in a large cohort of current IVDUs also will provide important contemporary data that will be of value to those involved in the prevention and treatment of drug abuse. HIV infection among IVDUs is a major and growing problem in the United States, affecting not only IVDUs themselves but also their sexual partners and children through perinatal transmission. Current data on HIV seroconversion among seronegative IVDUs are needed in order to design and evaluate the effects of interventions to prevent HIV infection and treat HIV-related disease in this population. Such data are not only important for observing effects of ongoing interventions such as bleach disinfection, condoms, and drug abuse treatment, but also to establish critical baseline information for the design of HIV vaccine studies.